@Article{,
author = {Ferguson, R. W., and Forbus, K. D.},
title = {GeoRep: A flexible tool for
spatial representation of line drawings},
journal = {Proceedings of the 18th
National Conference on Artificial Intelligence},
year = {2000},
OPTpages = {},
OPTnote = {Abstract:
A key problem in diagrammatic reasoning is understanding how people
reason about qualitative relationships in diagrams. We claim
that progress in diagrammatic reasoning is slowed by two
problems: (1) researchers tend to start from scratch, creating
new spatial reasoners for each new problem area, and (2)
constraints from human visual processing are rarely
considered. To address these problems, we created GeoRep, a
spatial reasoning engine that generates qualitative spatial
descriptions from line drawings. GeoRep has been successfully
used in several research projects, including cognitive
simulation studies of human vision. In this paper, we outline
GeoRep's architecture, explain the domain-independent and
domain-specific aspects of its processing, and motivate the
representations it produces. We then survey how GeoRep has been
used in three different projects-a model of symmetry, a model of
understanding juxtaposition diagrams of physical situations, and
a system for reasoning about military courses of action.},
OPTannote = {}
}
GeoRep takes as input a line drawing .fig file. If outputs the visual relations in it. Then it produces higher level relations.
GeoRep is a variant of the metric diagram/place vocabulary (MD/PV, Forbus, 1980).
The LLRD (low-level relational describer) describes grouping, proximity detection, reference frame relations, parallel lines, connection relations, ploygon and polyline detection, interval relations, and boundary descriptions. These are generated by universal routines (Ullman 1984) which operate independent of goal.
Primitive Shapes: line segments, circular arcs, circles, ellipses, splines, and text strings. Higher level relations (like arrows) can be programmed in as rules into the HLRD. Georep's content theory is in the LLRD, not the HLRD.
Items are proximate if their areas of influence overlap.
Place vocabularies are domain-specific visual relations.